• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.10
• /
• pp.75-80
• /
• 2017

In this paper, we describe the effect of the regularization method and the network with identity mapping on the performance of the activation functions in deep convolutional neural networks. The activation functions act as nonlinear transformation. In early convolutional neural networks, a sigmoid function was used. To overcome the problem of the existing activation functions such as gradient vanishing, various activation functions were developed such as ReLU, Leaky ReLU, parametric ReLU, and ELU. To solve the overfitting problem, regularization methods such as dropout and batch normalization were developed on the sidelines of the activation functions. Additionally, data augmentation is usually applied to deep learning to avoid overfitting. The activation functions mentioned above have different characteristics, but the new regularization method and the network with identity mapping were validated only using ReLU. Therefore, we have experimentally shown the effect of the regularization method and the network with identity mapping on the performance of the activation functions. Through this analysis, we have presented the tendency of the performance of activation functions according to regularization and identity mapping. These results will reduce the number of training trials to find the best activation function.

• Journal of the Korea Convergence Society
• /
• v.12 no.1
• /
• pp.29-35
• /
• 2021

License plate recognition plays a key role in intelligent transportation systems. Therefore, it is a very important process to efficiently detect the number and character areas. In this paper, we propose a method to effectively detect license plate number area by applying deep learning and semantic image segmentation algorithm. The proposed method is an algorithm that detects number and text areas directly from the license plate without preprocessing such as pixel projection. The license plate image was acquired from a fixed camera installed on the road, and was used in various real situations taking into account both weather and lighting changes. The input images was normalized to reduce the color change, and the deep learning neural networks used in the experiment were Vgg16, Vgg19, ResNet18, and ResNet50. To examine the performance of the proposed method, we experimented with 500 license plate images. 300 sheets were used for learning and 200 sheets were used for testing. As a result of computer simulation, it was the best when using ResNet50, and 95.77% accuracy was obtained.

• Journal of IKEEE
• /
• v.25 no.4
• /
• pp.664-671
• /
• 2021

This paper proposes a GPGPU structure that can reduce the number of operations and memory access by effectively applying a data reuse method to a convolutional neural network(CNN). Convolution is a two-dimensional operation using kernel and input data, and the operation is performed by sliding the kernel. In this case, a reuse method using an internal register is proposed instead of loading kernel from a cache memory until the convolution operation is completed. The serial operation method was applied to the convolution to increase the effect of data reuse by using the principle of GPGPU in which instructions are executed by the SIMT method. In this paper, for register-based data reuse, the kernel was fixed at 4×4 and GPGPU was designed considering the warp size and register bank to effectively support it. To verify the performance of the designed GPGPU on the CNN, we implemented it as an FPGA and then ran LeNet and measured the performance on AlexNet by comparison using TensorFlow. As a result of the measurement, 1-iteration learning speed based on AlexNet is 0.468sec and the inference speed is 0.135sec.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.4
• /
• pp.161-166
• /
• 2018

In this paper, we propose a method for the random selection of pooling operations for the regularization and reduction of cross validation in convolutional neural networks. The pooling operation in convolutional neural networks is used to reduce the size of the feature map and for its shift invariant properties. In the existing pooling method, one pooling operation is applied in each pooling layer. Because this method fixes the convolution network, the network suffers from overfitting, which means that it excessively fits the models to the training samples. In addition, to find the best combination of pooling operations to maximize the performance, cross validation must be performed. To solve these problems, we introduce the probability concept into the pooling layers. The proposed method does not select one pooling operation in each pooling layer. Instead, we randomly select one pooling operation among multiple pooling operations in each pooling region during training, and for testing purposes, we use probabilistic weighting to produce the expected output. The proposed method can be seen as a technique in which many networks are approximately averaged using a different pooling operation in each pooling region. Therefore, this method avoids the overfitting problem, as well as reducing the amount of cross validation. The experimental results show that the proposed method can achieve better generalization performance and reduce the need for cross validation.

• Journal of Multimedia Information System
• /
• v.7 no.1
• /
• pp.1-10
• /
• 2020

The image analysis is an important and predominant task for classifying the different parts of the image. The analysis of complex image analysis like histopathological define a crucial factor in oncology due to its ability to help pathologists for interpretation of images and therefore various feature extraction techniques have been evolved from time to time for such analysis. Although deep reinforcement learning is a new and emerging technique but very less effort has been made to compare the deep learning and deep reinforcement learning for image analysis. The paper highlights how both techniques differ in feature extraction from complex images and discusses the potential pros and cons. The use of Convolution Neural Network (CNN) in image segmentation, detection and diagnosis of tumour, feature extraction is important but there are several challenges that need to be overcome before Deep Learning can be applied to digital pathology. The one being is the availability of sufficient training examples for medical image datasets, feature extraction from whole area of the image, ground truth localized annotations, adversarial effects of input representations and extremely large size of the digital pathological slides (in gigabytes).Even though formulating Histopathological Image Analysis (HIA) as Multi Instance Learning (MIL) problem is a remarkable step where histopathological image is divided into high resolution patches to make predictions for the patch and then combining them for overall slide predictions but it suffers from loss of contextual and spatial information. In such cases the deep reinforcement learning techniques can be used to learn feature from the limited data without losing contextual and spatial information.

• Journal of Bio-Environment Control
• /
• v.29 no.4
• /
• pp.381-387
• /
• 2020

Various studies have been attempted to estimate and measure the fresh weight of crops. However, no studies have used raw images of sweet peppers to estimate fresh weight. Recently, image processing research using convolution neural network (CNN) that can use raw data is increasing. In this study, the crop fresh weight was estimated by using the images of sweet peppers as inputs of CNN. The experiment was performed in a greenhouse growing sweet pepper (Capsicum annuum L.). The fresh weight, the output of the CNN, was regressed based on the data collected through destructive investigation. The highest coefficient of determination (R²) of the trained CNN was 0.95. The estimated fresh weight showed a very similar trend to the actual measured value.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.5
• /
• pp.933-943
• /
• 2022

As the Internet develops and the utilization rate of computers increases, the threats posed by malware keep increasing. This leads to the demand for a system to automatically analyzes a large amount of malware. In this paper, an automatic malware analysis technique using a deep learning algorithm is introduced. Our proposed method uses CNN (Convolutional Neural Network) to analyze the malicious features represented as images. To reflect semantic information of malware for detection, our method uses the opcode frequency data of binary for image generation, rather than using bytes of binary. As a result of the experiments using the datasets consisting of 20,000 samples, it was found that the proposed method can detect malicious codes with 91% accuracy.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.2
• /
• pp.83-90
• /
• 2023

Blurred image is an important factor in lowering image recognition rates in Computer vision. This mainly occurs when the camera is unstablely out of focus or the object in the scene moves quickly during the exposure time. Blurred images greatly degrade visual quality, weakening visibility, and this phenomenon occurs frequently despite the continuous development digital camera technology. In this paper, it replace the modified building module based on the Deep multi-patch neural network designed with convolution neural networks to capture details of input images and Attention techniques to focus on objects in blurred images in many ways and strengthen the image. It measures and assigns each weight at different scales to differentiate the blurring of change and restores from rough to fine levels of the image to adjust both global and local region sequentially. Through this method, it show excellent results that recover degraded image quality, extract efficient object detection and features, and complement color constancy.

• East Asian Economic Review
• /
• v.28 no.3
• /
• pp.359-388
• /
• 2024

With the escalation of global trade, the Chinese commodity futures market has ascended to a pivotal role within the international shipping landscape. The Shanghai Containerized Freight Index (SCFI), a leading indicator of the shipping industry's health, is particularly sensitive to the vicissitudes of the Chinese commodity futures sector. Nevertheless, a significant research gap exists regarding the application of Chinese commodity futures prices as predictive tools for the SCFI. To address this gap, the present study employs a comprehensive dataset spanning daily observations from March 24, 2017, to May 27, 2022, encompassing a total of 29,308 data points. We have crafted an innovative deep learning model that synergistically combines Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) architectures. The outcomes show that the CNN-LSTM model does a great job of finding the nonlinear dynamics in the SCFI dataset and accurately capturing its long-term temporal dependencies. The model can handle changes in random sample selection, data frequency, and structural shifts within the dataset. It achieved an impressive R² of 96.6% and did better than the LSTM and CNN models that were used alone. This research underscores the predictive prowess of the Chinese futures market in influencing the Shipping Cost Index, deepening our understanding of the intricate relationship between the shipping industry and the financial sphere. Furthermore, it broadens the scope of machine learning applications in maritime transportation management, paving the way for SCFI forecasting research. The study's findings offer potent decision-support tools and risk management solutions for logistics enterprises, shipping corporations, and governmental entities.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.10
• /
• pp.97-104
• /
• 2024

This paper proposes a method for classifying the grade of stainless steel using a portable NIR(Near Infrared Ray) spectrometer and a CNN(Convolutional Neural Network) deep learning model. Traditionally, methods for classifying stainless steel grades have included chemical analysis, magnetic testing, molybdenum spot tests, and portable XRF devices. In addition, a classification method using a machine learning model with element concentration and heat treatment temperature as parameters was presented in the paper. However, these methods are limited in their application to everyday products, such as kitchenware and cookware, due to the need for reagents, specialized equipment, or reliance on professional services. To address these limitations, this paper proposes a simple method for classifying the grade of stainless steel using a NIR spectrometer and a CNN model. If the method presented in this paper is installed on a portable device as an on-device in the future, it will be possible to determine the grade of stainless steel used in the product, and to determine on-site whether a product made of low-cost material has been disguised as a high-cost product.